diff --git a/images/DML_model_indicator.PNG b/images/DML_model_indicator.PNG new file mode 100644 index 00000000..bb999511 Binary files /dev/null and b/images/DML_model_indicator.PNG differ diff --git a/images/QPS_empty_batches.PNG b/images/QPS_empty_batches.PNG new file mode 100644 index 00000000..fd295720 Binary files /dev/null and b/images/QPS_empty_batches.PNG differ diff --git a/images/RPS_client.PNG b/images/RPS_client.PNG new file mode 100644 index 00000000..aa9b19f5 Binary files /dev/null and b/images/RPS_client.PNG differ diff --git a/images/TPS_event.PNG b/images/TPS_event.PNG new file mode 100644 index 00000000..62416ba0 Binary files /dev/null and b/images/TPS_event.PNG differ diff --git a/images/TPS_rows.PNG b/images/TPS_rows.PNG new file mode 100644 index 00000000..d84b94cd Binary files /dev/null and b/images/TPS_rows.PNG differ diff --git a/images/TPS_transaction.PNG b/images/TPS_transaction.PNG new file mode 100644 index 00000000..9d0f2344 Binary files /dev/null and b/images/TPS_transaction.PNG differ diff --git a/images/client_band.PNG b/images/client_band.PNG new file mode 100644 index 00000000..40024bcf Binary files /dev/null and b/images/client_band.PNG differ diff --git a/images/store_remain_events.PNG b/images/store_remain_events.PNG new file mode 100644 index 00000000..4e617a21 Binary files /dev/null and b/images/store_remain_events.PNG differ diff --git a/images/store_remain_memory.PNG b/images/store_remain_memory.PNG new file mode 100644 index 00000000..c8e6032f Binary files /dev/null and b/images/store_remain_memory.PNG differ diff --git a/prometheus/README.md b/prometheus/README.md deleted file mode 100644 index 1a76709b..00000000 --- a/prometheus/README.md +++ /dev/null @@ -1,89 +0,0 @@ -###Canal Prometheus 简介 - -Canal server 性能指标监控基于prometheus的实现。 - -关于prometheus，参见[官网](https://prometheus.io/) - -#### 当前监控的canal相关原始指标列表： - -| 指标 | 说明 | 单位 | 精度 | -| :---- | :---- | ----: |----: | -|canal_instance_traffic_delay |instance延迟 | 毫秒 | 毫秒 | -|canal_instance_transactions |instance接收transactions计数|-|-| -|canal_instance_row_events |instance接收rowData类型events计数|-|-| -|canal_instance_rows_counter |instance接收events包含变更行数计数|-|-| -|canal_instance |instance基本信息|-|-| -|canal_instance_subscriptions |instance订阅数量|-|-| -|canal_instance_publish_blocking_time |instance dump线程publish阻塞时间(仅parallel解析模式)|毫秒|纳秒| -|canal_instance_received_binlog_bytes |instance接收binlog字节数|byte|-| -|canal_instance_parser_mode |instance解析模式(是否开启parallel解析)|-|-| -|canal_instance_client_packets |instance client请求packets计数|-|-| -|canal_instance_client_bytes|向instance client发送数据包字节计数|byte|-| -|canal_instance_client_empty_batches|向instance client发送的空batch计数|-|-| -|canal_instance_client_request_error|instance client请求失败计数|-|-| -|canal_instance_client_request_latency |instance client请求延迟概况|-|-| -|canal_instance_sink_blocking_time |instance sink线程put数据至store的阻塞时间|毫秒|纳秒| -|canal_instance_store_produce_seq |instance store接收到的events sequence number|-|-| -|canal_instance_store_consume_seq |instance store成功消费的events sequence number|-|-| -|canal_instance_store |instance store基本信息|-|-| -|canal_instance_store_produce_mem |instance store接收到的所有events占用内存总量|byte|-| -|canal_instance_store_consume_mem |instance store成功消费的所有events占用内存总量|byte|-| - -#### JVM 相关信息 -> The Java client includes collectors for garbage collection, memory pools, JMX, classloading, and thread counts. These can be added individually or just use the DefaultExports to conveniently register them. -> >DefaultExports.initialize(); - -详见：[prometheus/client_java](https://github.com/prometheus/client_java) - - -#### Quick start - -1. 安装并部署对应平台的prometheus，参见[官方guide](https://prometheus.io/docs/introduction/first_steps/) - -2. 配置prometheus.yml，添加canal的job，示例: -``` - - job_name: 'canal' - static_configs: - - targets: ['localhost:11112'] //端口配置即为canal.properties中的canal.metrics.pull.port -``` - -3. 启动prometheus与canal server - -#### Prometheus expression 与相应 use case -* Sink线程空闲比 -``` -clamp_max(rate(canal_instance_sink_blocking_time{destination="example"}[2m]), 1000) / 10 -``` -**sink线程idle时间片比例(向store中put events时)。若idle占比很高，则store总体上处于满的状态，client的consume速度低于server的produce速度** - -_简单说明一下range-vector,通俗来说表达式会用时间点前range-vector period内的所有samples参与运算。_ -_range-vector值如果太小，图会碎片化；反之，实时性会比较差。请结合scrape_interval设定合理的值。_ - -* Dump线程空闲比 -``` -clamp_max(rate(canal_instance_publish_blocking_time{destination="example"}[2m]), 1000) / 10 -``` -**dump线程idle时间片比例(仅parallel mode, dump线程向disruptor发布event时)。若idle占比较高：** - -**1. Sinking idle ratio也很高，则总体还是因为client的consume速度相对较慢。** - -**2. Sinking idle ratio较低，那么server端parser是性能瓶颈，可参考[Performance](https://github.com/alibaba/canal/wiki/Performance)进行tuning.** - -* Delay(seconds) -``` -canal_instance_traffic_delay / 1000 -``` -**Instance binlog消费延迟，有两个注意点：** - -**1. 如果Canal已经消费至最新position，且binlog长时间未更新，delay的resolution受到master_heartbeat_period的影响，目前频率为15秒。** - -**2. Limitations: 如果store满了(恰好store中的数据包含最新的position，且MySQL binlog停止更新)，且client暂停消费，那么delay会不断增长。当然，满足这些条件的概率极低。** - -* Binlog接收速率(KB/s) -``` -rate(canal_instance_received_binlog_bytes{destination="example"}[2m]) / 1024 -``` -**'Sink线程空闲比'与'Dump线程空闲比'都很低，delay还是很高的情况，请查看binlog接收速率是否符合预期。** - - ---To be continued... diff --git a/server/src/main/java/com/alibaba/otter/canal/server/netty/listener/ChannelFutureAggregator.java b/server/src/main/java/com/alibaba/otter/canal/server/netty/listener/ChannelFutureAggregator.java index b962b9a0..71051db0 100644 --- a/server/src/main/java/com/alibaba/otter/canal/server/netty/listener/ChannelFutureAggregator.java +++ b/server/src/main/java/com/alibaba/otter/canal/server/netty/listener/ChannelFutureAggregator.java @@ -7,6 +7,7 @@ import org.jboss.netty.channel.ChannelFuture; import org.jboss.netty.channel.ChannelFutureListener; import static com.alibaba.otter.canal.server.netty.CanalServerWithNettyProfiler.profiler; +import static com.alibaba.otter.canal.server.netty.NettyUtils.HEADER_LENGTH; /** * @author Chuanyi Li @@ -32,7 +33,7 @@ public class ChannelFutureAggregator implements ChannelFutureListener { .destination(destination) .type(type) .request(request) - .amount(amount) + .amount(amount + HEADER_LENGTH) .latency(latency) .errorCode(errorCode) .empty(empty)